UK Banks Remain Blocked from Anthropic's Mythos, Says Bailey

Bank of England Governor Andrew Bailey confirmed at a central banks conference in Iceland that UK financial institutions remain blocked from accessing Mythos, a powerful AI model from Anthropic. This restriction significantly hampers banks' efforts to use cutting-edge technologies to protect their digital infrastructure.

Ban on Access to Advanced Model

Mythos is considered one of the most powerful AI systems ever developed by Anthropic. However, UK banks remain cut off from using it. The reasons for this restriction are not fully disclosed, but according to Bailey, the decision is linked to regulators' cautious approach to implementing new AI tools in critical sectors of the economy. The Bank of England requires additional analysis and risk assessment before allowing new AI models to work with financial infrastructure. This restriction appears paradoxical given that cyberattacks on banks are becoming increasingly sophisticated and dangerous. Financial institutions need better tools to protect their systems, but cannot use the most advanced solutions available on the global market.

Alternatives Banks Are Using

According to Bailey, UK banks are not remaining passive. They are actively testing various alternative AI models and tools for checking and strengthening their cybersecurity systems:

• Commercial models from OpenAI and other vendors
• Open-source solutions available to developers
• In-house proprietary developments for security analysis
• Specialized platforms from cybersecurity companies
• Traditional penetration testing and manual audit methods

However, all these alternatives fall short of Mythos in power and capabilities. Many banks are forced to use multiple tools simultaneously to achieve the necessary level of protection. This requires substantial time and financial expenditures, slowing modernization and diverting resources.

Widening Gap Between Innovation and Regulation

The Mythos situation reflects a broader trend: regulators are moving slower than technology is developing. The Bank of England, like other central banks, fears uncontrolled implementation of AI in critical systems. On one hand, this is reasonable—financial security requires reliability and predictability. On the other hand, excessive caution may slow innovation and weaken the competitiveness of the UK financial sector. The question of access to Mythos will become key in negotiations between the Bank of England, Anthropic, and financial institutions. If the restriction persists, it could accelerate the migration of operations to other jurisdictions.

What This Means

The Mythos situation demonstrates a growing gap between technical capabilities and regulatory caution. For the UK financial sector, this means a real risk of falling behind in the race for the best digital tools. At the same time, it highlights the need to find the right balance between innovation and stability in the financial system.